Search

This paper reports on the implementation and characterization of a hermetically-sealed single-proof-mass multi-directional accelerometer contact microphone exhibiting wide operational bandwidth (>10kHz) and low noise (Brownian noise

Pandemics such as Covid-19 change the status quo. The things we take for granted become no longer true. When the disease is contagious then in-person communication is fraught with danger for all parties. Despite all these dangers people still required to meet in person for various reasons that could range from personal to official to government business. Can we have a way to understand the risk of a person being infected as compared to another person so that we can make decisions of segregating the two people or to decline entry to a person? In this paper we provide an architecture and an approach that caters to estimating a score for each person based on GPS (Global Positioning System) locations. A score is computed by an application running in cloud and receiving as input the GPS trajectory information from each person's device (like a mobile or a smart watch). With the scores of any two people in hand one can predict who is more likely to be infected. We complement this approach with extensive simulation results to validate our approach. Our results show that we can achieve a high accuracy (80% to 90%) of predicting which person (of the two being compared) has an infection.

Every year, the crop output of India is significantly affected due to delayed detection of diseases in crops. This research is a contribution to the farmers in their battle against coffee plants diseases. It will help in timely detection of diseases, resulting in increased coffee production output of India. Many coffee plant diseases like Leaf Rust, Cercospora Spots have clear visual symptoms and thus can be extracted and their classification can be done. Convolutional Neural Networks (CNNs) has proved its efficiency and accuracy in the field of image classification and pattern recognition. Hence it can act as a powerful tool in the diagnosis of coffee leaves diseases since these symptoms have clearly distinguishable patterns. Thus, a Convolution Neural Network model is proposed which utilizes the technique of Transfer Learning, reducing the training time taken by the model significantly. Further, to achieve a higher success rate, Data Augmentation technique is applied to enlarge the dataset used to train the network. The proposed model has achieved a high accuracy of 97.61%.

This paper presents progress on the development of a high-performance single-chip timing and inertial measurement unit (TIMU) based on environmentally-robust high-frequency micromechanical devices and the hermetically-sealed HARPSS+ silicon MEMS process. The fabrication platform provides the unique capability of producing deep-sub-micron capacitive gaps (i.e. nano-gaps) in various orientations, including 54°-slanted electrodes, enabling low-voltage mode-matching and alignment of triaxial high-frequency resonator gyroscopes with high sensitivity. The resonator (i.e. mode-matched) gyroscopes and quasi-static accelerometers are sealed in separate cavities with different pressure levels through a single bonding step, which allows for the co-fabrication and co-packaging of high-dynamic-range high-Q bulk acoustic wave (BAW) resonator 3-axis gyroscopes and critically-damped high-bandwidth (above 10kHz) micro-gravity 3-axis accelerometers. High-frequency and high-Q nano-gap timing resonators with low motional impedance are fabricated on the same substrate with the IMU, which favors low-jitter high-frequency oscillator implementations and enables OCXO demonstrations with sub-ppm stability. Fabricated devices were tested against wafer thinning process, high shocks, aging effects and extensive temperature variations, verifying the robustness of the designs and fabrication process. The high performance, high robustness, and small form factor make the HARPSS+ precision TIMUs an attractive solution for demanding inertial navigation and guidance applications.

We report on a novel wafer-level-packaged accelerometer contact microphone to pick up acoustic vibrations orthogonal to its contact surface without responding to airborne acoustic emissions. The design uses a high-bandwidth low-noise unidirectional capacitive accelerometer with a deformable frame suspending a proof-mass, designed to have larger displacement at the edge to allow for the implementation of shock stops. The sense electrodes use a nano-gap to provide high electrical sensitivity with high stiffness design and nano-gap damping electrodes enable near-critical damping in a low vacuum encapsulation to achieve a high bandwidth above 14 kHz. We demonstrate that the miniature accelerometer contact microphone placed on the throat can record human speech and heart sounds with high fidelity while rejecting surrounding noises.

In this paper, we have done an assessment to evaluate the efficiency of different feature descriptor algorithms like Local Binary Pattern (LBP), Scale Invariant Feature Transform (SIFT), Speeded-Up Robust Features (SURF) and Oriented FAST and Rotated BRIEF (ORB) on gender classification problem. The evaluation is done with and without using the feature descriptors and the results are compared to get a better understanding of the algorithms. Experiments are carried out on CAS-PEAL FRONTAL database.

This paper presents a single-chip timing and inertial measurement unit (TIMU) with high-frequency modematched gyroscopes, high-bandwidth accelerometers, and a high-frequency high-Q bulk acoustic wave (BAW) timing resonator integrated on a 4.5mm∗5.5mm platform using the HARPSS+ process, which enables vertical, horizontal, and 54°-slanted electrodes with scalable nano-gaps. The TIMU is the first single-chip IMU featuring 3-axis high-frequency quadrature-compensated resonant (i.e. mode-matched) gyroscopes. The 3-axis gyroscopes have high shock and vibration resistance and present sub-1°/√h angle random walk (ARW) and ∼10°/h bias instability (BI) limited by discrete electronics. The 3-axis accelerometers show sub-300μg/√Hz velocity random walk (VRW) and sub-100μg BI. The timing resonator has a high Q of 47k with a high frequency of 85MHz. Further improvements can be achieved with optimized interface electronics to meet high-end application requirements.

The system is specially developed for elderly people living alone in their homes. A novel method for detection of fall in home environment is presented. Video clips of certain duration specifying fall and non-fall activities are considered for experimentation. Shi-Tomasi algorithm is used to find interest points which are tracked by Pyramidal Lucas-Kanade algorithm. The speed and direction of motion is computed along with maximum displacement of interest point. Also the last positions of interest points were observed to declare the fall activity. The results show 95% accuracy for non-fall activities and 96.67% for fall activities with UR fall dataset.

We consider the issue of call-center scheduling coupled with capacity planning of both service agents (servers) and physical seats (space) in an environment where call arrival rates are highly variable and the call center is subjected to a global service level constraint. The paper is motivated by the work with the business process outsourcing unit providing technical support services from growth markets like India where real-estate is one of the major cost drivers along with human capital. Our paper has two distinctive features. Firstly, we model capacity planning problem for seats and scheduling of agents as one single optimization problem leveraging variable service rates of human agents; information which is ignored by workforce planning teams while modeling. Secondly, we formulate the problem as a mixed integer linear programming (MILP) problem which is not trivial given the non-linear nature of service level constraint. We show that the set of linear inequalities required to represent service level constraint for each period can be generated cheaply. We conduct extensive experimentation to compare solutions of the MILP formulation recognizing variability in service rate with the traditional formulation based on mean valued service rates. We find that, in general, our formulation provides a significant reduction in space requirement and the solution minimizes the over-achievement of service level requirement.

An IT services businesses must continually manage the assignment and movement of practitioners to or between projects balancing multiple objectives (1) to minimize their idle bench, while aiming (2) to increase revenue from new project opportunities. The key to simultaneously achieving these objectives is a well managed Service System for Workforce Deployment with distributed decision making. In this paper, we discuss the decision support needs for this Service System, and present a decision support tool that computes an 'optimal' assignment of practitioners to projects,and improves the effectiveness and efficiency of this assignment process. The tool comprises of a matching module that matches practitioners to project openings, an optimization module that produces globally optimal recommendations, and a user interface for displaying the recommendations. We describe the decision support tool, our experiences with developing and deploying it in an IT services company, and our efforts to promote a common shared perspective among decision-makers enabling multiple decision makers to independently arrive at near-optimal decisions using the tool recommendations.

Rapidly growing service delivery organizations need efficient tools to manage their business and reduce the cost of growth. These delivery businesses run round the clock due to demands coming from different geographical locations for different shifts. Cost of adding new physical infrastructure and space to tap incoming business is a key inhibitor of growth. Physical space availability and its effective management is the key infrastructure enabler for effective business delivery apart from human resources. A considerable impact can be realized if space utilization can be increased by optimal allocation around different work shifts. Optimal space utilization can bring down the number of physical seats required to serve the existing demand and also allow service delivery organizations to commit to new business needs with the existing capacity. The paper addresses this business problem and proposes an effective seat utilization planning solution using a mathematical programming approach to meet various strategic and tactical business objectives in this setting.